Statistics and Probability Commons^™

Markov Chain Monte Carlo Bayesian Predictive Framework For Artificial Neural Network Committee Modeling And Simulation, Michael S. Goodrich

Computational Modeling & Simulation Engineering Theses & Dissertations

A logical inference method of properly weighting the outputs of an Artificial Neural Network Committee for predictive purposes using Markov Chain Monte Carlo simulation and Bayesian probability is proposed and demonstrated on machine learning data for non-linear regression, binary classification, and 1-of-k classification. Both deterministic and stochastic models are constructed to model the properties of the data. Prediction strategies are compared based on formal Bayesian predictive distribution modeling of the network committee output data and a stochastic estimation method based on the subtraction of determinism from the given data to achieve a stochastic residual using cross validation. Performance for Bayesian …

A Metaevaluation Of Energy Efficiency Evaluations, Brandy Brown

Dissertations

This study systematically reviews the methodological characteristics of energy efficiency evaluations and uses metaevaluation to assess its quality. Metaevaluation is used to systematically assess the quality of evaluation products, confirm that evaluations deliver sound findings and conclusions, are useful to the client, are credible, are ethically conducted, and are done as cost-effective as possible. The results of this study show that the ability to accurately assess evaluation for methodological quality using evaluations reports as a primary data source depends on the presence of detailed descriptions of evaluation methods. Furthermore, the study suggests that methodological variations of energy efficiency evaluations coalesce …

A Metaevaluation Of Evaluations Of Health Care Programs That Employ The Chronic Care Model, Jan Fields

Dissertations

Background: The purpose of this dissertation is to explore the use of metaevaluation to evaluate the quality of healthcare studies conducted on programs that employ the Chronic Care Model (CCM) to provide chronic illness care. In this study, healthcare studies of CCM programs are regarded as program evaluations. Method: Using a non-experimental cross-sectional design, 28 healthcare studies of CCM programs were evaluated using the accuracy standards portion of the Program Evaluations Metaevaluation Checklist (Stufflebeam, 2011). The results of the metaevaluations were analyzed and compared to the HEAL grade of the same healthcare studies as determined by the Hierarchy of Evidence …

Using Multi-Objective Value Estimation To Support Predictive Analytics For Human Service Project Management, David D. Wingard

Dissertations

Human service organizations need outcome measurement approaches that support project management for efficiency and effectiveness. While, in recent years, human services have increased their capacity to manage data and measure outcomes empirically, several barriers remain. First, current outcome measurement practices are not designed to effectively support the management of human services programs for maximum efficiency and effectiveness. Second, human services organizations need a methodology to manage programs to identified outcomes. This dissertation explored meaningful solutions to both issues. In Paper 1 (Chapter II), this dissertation assessed strengths and limitations of current outcome evaluation approaches and suggested an innovative application of …

Harnessing Complexity: Analysis Methodology And Ethical Framework To Facilitate Utilization Of Video Data In Evaluations, Kurt A. Wilson

Dissertations

Most evaluations in the nonprofit and international development sectors are conducted in contexts of complexity; the specific intervention being evaluated is but one of many interrelated factors influencing the desired outcome. Video data, especially when directly generated by program participants, can provide both exceptionally rich qualitative data as well as contextually-relevant feedback within complex systems. Despite these unique strengths and opportunities, video data is underutilized in the field of evaluation. This dissertation addresses specific barriers associated with video data through three inter-related papers: Papers one and two (Chapters II and III) present the findings from two interrelated studies of an …

Multivariate Autoregressive Time Series Using Schweppe Weighted Wilcoxon Estimates, Jaime Burgos

Dissertations

The increasing needs of forecasting techniques has led to the popularity of the vector autoregressive model in multivariate time series analysis, which has become of typical use across different fields due to its simplicity in application. The traditional method for estimating the model parameters is the least squares minimization, due to the linear nature of the model and its similarity with multivariate linear regression. However, since least squares estimates are sensitive to outliers, more robust techniques have become of interest. This manuscript investigates a robust alternative by obtaining the estimates using a weighted Wilcoxon dispersion with Schweppe-type weights. The first …

Improving The Design Of Cluster-Randomized Trials In Education: Informing The Selection Of Variance Design Parameter Values For Science Achievement Studies, Carl D. Westine

Dissertations

The purpose of this three-essay dissertation is to provide practical guidance to evaluators planning cluster-randomized trials (CRTs) of science achievement. In an educational setting, interventions are often administered at the cluster level, while outcomes are typically measured at the student level through standardized achievement testing. When evaluating an intervention, a CRT is appropriate because it allows for treatment to be modeled at a different level than the unit of analysis, and properly accounts for the violation of independence that occurs due to nesting. Accurately designing a CRT involves estimating variance parameters (i.e., intraclass correlations [ICCs] and percent of variance explained …

Bounds To Evaluate The Pure/Natural Direct Effect Without Cross-World Counterfactual Independence, Eric Tchetgen Tchetgen, Kelesitse Phiri

Harvard University Biostatistics Working Paper Series

No abstract provided.

A Global Partial Likelihood Estimator Of The Time-Varying Effects For Time-Dependent Treatment, Huazhen Lin, Zhe Fei, Yi Li

The University of Michigan Department of Biostatistics Working Paper Series

The timing of time-dependent treatment - e.g., when to perform kidney transplantation - is an important factor for evaluating treatment efficacy. A naive comparison between the treatment and nontreatment groups, while ignoring the timing of treatment, typically yields results that might biasedly favor the treatment group, as only patients who survive long enough will get treated. On the other hand, studying the effect of time-dependent treatment is often complex, as it involves modeling treatment history and accounting for the possible time-varying nature of the treatment effect. We propose a varying-coefficient Cox model that investigates the efficacy of time-dependent treatment by …

A Unification Of Mediation And Interaction: A Four-Way Decomposition, Tyler J. Vanderweele

Harvard University Biostatistics Working Paper Series

It is shown that the overall effect of an exposure on an outcome, in the presence of a mediator with which the exposure may interact, can be decomposed into four components: (i) the effect of the exposure in the absence of the mediator, (ii) the interactive effect when the mediator is left to what it would be in the absence of exposure, (iii) a mediated interaction, and (iv) a pure mediated effect. These four components, respectively, correspond to the portion of the effect that is due to neither mediation nor interaction, to just interaction (but not mediation), to both mediation …

Five Fundamental Gaps In Nature-Nurture Science, Peter J. Taylor

Working Papers on Science in a Changing World

Difficulties identifying causally relevant genetic variants underlying patterns of human variation have been given competing interpretations. The debate is illuminated in this article by drawing attention to the issue of underlying heterogeneity—the possibility that genetic and environmental factors or entities underlying a trait are heterogeneous—as well as four other fundamental gaps in the methods and interpretation of classical quantitative genetics: "Genetic" and "environmental" fractions of variation in traits are distinct from measurable genetic and environmental factors underlying the traits’ development; Standard formulas for partitioning variation in human traits are unreliable; Methods for translation from fractions of variation to measurable …

New Statistical Methods For Analysis Of Historical Data From Wildlife Populations, Trevor Hefley

Department of Statistics: Dissertations, Theses, and Student Research

Wildlife biologists, many times with the help of ordinary citizens, have developed and maintained long-term datasets for monitoring the status of wildlife populations. These datasets can range from a collection of citizen-reported sightings of a rare species, to datasets collected by biologists using standardized methods. The commonality is that these datasets span a temporal and spatial scale that is beyond the scope of most scientific studies. Ensuring the continued persistence of wildlife populations requires predictions of the impact of human actions. Regardless if the predictions are quantitative or qualitative, the best we can do is use the past data to …

New Malignancies After Squamous Cell Carcinoma And Melanomas: A Population-Based Study From Norway, Trude E. Robsahm, Margaret R. Karagas, Judy R. Rees, Astri Syse

Dartmouth Scholarship

Skin cancer survivors experience an increased risk for subsequent malignancies but the associated risk factors are poorly understood. This study examined the risk of a new primary cancer following an initial skin cancer and assessed risk factors associated with second primary cancers.

Efficiently Identifying Failures Using Quantitative Tests, Matrix-Pooling And The Em-Algorithm, Brett Hanscom, Susanne May, Jim Hughes

UW Biostatistics Working Paper Series

Pooled-testing methods can greatly reduce the number of tests needed to identify failures in a collection of samples. Existing methodology has focused primarily on binary tests, but there is a clear need for improved efficiency when using expensive quantitative tests, such as tests for HIV viral load in resource-limited settings. We propose a matrix-pooling method which, based on pooled-test results, uses the EM algorithm to identify individual samples most likely to be failures. Two hundred datasets for each of a wide range of failure prevalence were simulated to test the method. When the measurement of interest was normally distributed, at …

A Predictive Enrichment Procedure To Identify Potential Responders To A New Therapy For Randomized, Comparative, Controlled Clinical Studies, Junlong Li, Lihui Zhao, Lu Tian, Tianxi Cai, Brian Claggett, Andrea Callegaro, Benjamin Dizier, Bart Spiessens, Fernando Ulloa-Montoya, L. J. Wei

Harvard University Biostatistics Working Paper Series

No abstract provided.

Asymmetric Empirical Similarity, Joshua C. Teitelbaum

Georgetown Law Faculty Publications and Other Works

The paper offers a formal model of analogical legal reasoning and takes the model to data. Under the model, the outcome of a new case is a weighted average of the outcomes of prior cases. The weights capture precedential influence and depend on fact similarity (distance in fact space) and precedential authority (position in the judicial hierarchy). The empirical analysis suggests that the model is a plausible model for the time series of U.S. maritime salvage cases. Moreover, the results evince that prior cases decided by inferior courts have less influence than prior cases decided by superior courts.

The Association Of Calcium Intake And Other Risk Factors With Cardiovascular Disease Among Obese Adults In Usa, Yang Chen, Sheryl Strasser, Katie Callahan, David Blackley, Yan Cao, Liang Wang, Shimin Zheng

ETSU Faculty Works

In this study, we used a cross-sectional study design to examine the relationship between the calcium intake and risk factors for CVD among obese adults by using continuous waves of National Health and Nutrition Examination Survey (NHANES) data 1999-2010. The association between calcium intake and risk factors of CVD (hypertension, total cholesterol, HDL, glycohemoglobin), CRP, albuminuria) is assessed among obese adults in USA. The incidence of Cardiovascular Disease (CVD) is high among obese people. The potential effects of inadequate calcium intake on CVD are receiving increased epidemiologic attention. Understanding the association between risk factors for CVD and calcium intake among …

Probabilistic Modeling Of Lava Flows: A Hazard Assessment For The San Francisco Volcanic Field, Arizona, Aleeza Harburger

USF Tampa Graduate Theses and Dissertations

This study serves as a first step towards a comprehensive hazard assessment for the San Francisco volcanic field in northern Arizona, which can be applied to local response plans and educational initiatives. The primary goal of this thesis is to resolve the conditional probability that, given a lava flow effusing from a new vent in the San Francisco volcanic field, it will inundate the city limits of Flagstaff. The spatial distribution of vents within the San Francisco volcanic field was analyzed in order to execute a lava flow simulation to determine the inundation hazard to Flagstaff. The Gaussian kernel function …

A Novel Three Serum Phospholipid Panel Differentiates Normal Individuals From Those With Prostate Cancer, Nima Patel, Robert L. Vogel, Kumar Chandra-Kuntal, Wayne Glasgow, Uddhav Kelavkar

Biostatistics: Faculty Publications

Background: The results of prostate specific antigen (PSA) and digital rectal examination (DRE) screenings lead to both under and over treatment of prostate cancer (PCa). As such, there is an urgent need for the identification and evaluation of new markers for early diagnosis and disease prognosis. Studies have shown a link between PCa, lipids and lipid metabolism. Therefore, the aim of this study was to examine the concentrations and distribution of serum lipids in patients with PCa as compared with serum from controls.

Method: Using Electrospray ionization mass spectrometry (ESI-MS/MS) lipid profiling, we analyzed serum phospholipids from age-matched subjects who …

Clustering Survival Outcomes Using Dirichlet Process Mixture, Lili Zhao, Jingchunzi Shi, Tempie H. Shearon, Yi Li

The University of Michigan Department of Biostatistics Working Paper Series

Motivated by the national evaluation of mortality rates at kidney transplant centers in the United States, we sought to assess transplant center long- term survival outcomes by applying a methodology developed in Bayesian non-parametrics literature. We described a Dirichlet process model and a Dirichlet process mixture model with a Half-Cauchy for the estimation of the risk- adjusted effects of the transplant centers. To improve the model performance and interpretability, we centered the Dirichlet process. We also proposed strategies to increase model's classification ability. Finally we derived statistical measures and created graphical tools to rate transplant centers and identify outlying centers …